Software Testing through Evidence Gathering

Transcription

1 International Journal of Information Computation Technology. ISSN Volume 3, Number 7 (2013), pp International Research Publications House irphouse.com /ijict.htm Software through Evidence Gathering Priyanka Mathur #1 Swati V. Che *2 1 Department of Computer Science, The IIS University, Jaipur, INDIA 2 International School of Informatics Management, Department of Computer Science, Jaipur, INDIA. Abstract An Evidence-based approach is using a best available evidence for making a thoughtful decision about a given set of problem. Evidencebased approach is an amalgamation of individually gained expertise with the evidences gathered through an organized research based approach. Its basic principles are that all practical decisions made should 1) be based on research studies 2) that these research studies are selected interpreted according to some specific norms characteristic for Evidence Based Practice[EBP]. Many software testing techniques are proposed to test various types of software based upon the evidences gathered an effective testing methodology is adopted for a software. This is evidence based approach for software testing. The software techniques are classified on the basis of three criteria s a) stepwise code reading b) performing functional testing by adopting boundary value analysis c) all the statements are covered using structural testing. The study compares the strategies with respect to fault detection effectiveness fault detection time complexity. In Evidence-based software engineering (EBSE), all the experiences are properly documented in order to inform software practice adoption decisions. In this research paper, the study factor would be the technology of interest. The technological specifications should be very detailed not at a very high level of abstraction that is the software lifecycle all the design methods should be properly read documented only then should the engineer collect evidences on it design the software generation model. In this paper we analyze the gathered evidence so as to classify the testing strategies on the basis of applicability types of testing.

2 626 Priyanka Mathur et al Various software testing strategies were studied in which white box testing structural testing are the most preferred methods when using GA (Genetic Algorithm) SA(Simulated Annealing) as a technique. 1. Introduction Meeting the objective of the paper software testing techniques identified are Fig. 1: Techniques. In accordance with the aforesaid objectives, research papers from eminent researchers were studies Basili (1990) suggested that for the success of a software product software testing fault detection activities should be exactly adequately understood, as they are very crucial for the success of a software product. Thus an experimentation methodology is applied to test the software effectively. Some common testing techniques are applied to different types of software software testing effectiveness is measured on the basis of several factors like: technique Software type Fault type Tester experience And an intercommunication among all these factors. The most commonly referred software testing techniques are Functional testing (black box approach) Structural testing (white box approach) Code reading In Code reading identification of subprograms, evaluation of their functionality is done. Further all subprograms are integrated again their functionality is tested.

3 Software through Evidence Gathering 627 Bergstra (2012) suggested a new software testing strategy: Instruction Sequence testing. In the paper the researcher compares testing from the point of view of two different classical definitions of testing. The first definition by King (1976) says that in testing a small sample of data that the program is expected to hle program is presented to the program. If the program is judged to produce correct results for the sample it is assumed to be correct. And the second definition as per Singh (2012) is is the process of executing a program with the intent of finding faults. A comparative analysis of the first theory reflects that it s a program working on a machine that produces output, there may or may not be any human intervention in testing. The second definition reflects that a test is successful if only if it finds faults. The author crafts a term Polinseq which means polyadic Instruction Sequence testing wherein a program is tested instruction wise, marking a difference between program testing software testing. The complexity of the program program testing makes Polinseq as a good testing technique but not a risk free technique. Harman (2009) ET al. is of the opinion that software testing is the process to measure the quality of developed software. Quality here encompasses completeness, correctness, security other non functional requirements like reliability, capability, maintainability, efficiency, portability, compatibility usability. Evidence based software testing holds a good weightage in this field more than half the percentage papers are published regarding software testing. Evidence based or Search based test data generation is the possible number of inputs to the program or test cases their fitness function. As an example taken in this report to achieve branch the fitness function accesses the closeness of test to executing an uncovered branch; in order to find worst case execution time, fitness is simply the duration of execution for the test case in question. Gathering data from the research paper presented by Khan(2010) the conclusions drawn are that: The main aim of testing can be quality assurance, validation verification. Automated testing can be performed in well controlled softwares. A successful testing technique uncovers an undiscovered error. Generally software testing is done to affirm the quality of software by systematically testing the software in controlled circumstances. The testing techniques listed above are suitable for Object Oriented Paradigms. There has been much other work on structural test data generation for the OO paradigm. And to analyze the gathered evidence a comparative study of objective/fitness function problems faced in are listed in Table 1

4 628 Priyanka Mathur et al technique Structural Structural White-box testing Structural Table 1: Comparative study of objective/fitness function. Technique Description Objective /Fitness function GA, SA (Simulated annealing) GA GA Genetic algorithms (GAs) evolutionar y strategies (ESs) Branch, data flow,decision. Combine nonfunctional testing goals with based adequacy criterion as a multi objective problem. Test data generation Test data generation path branch dynamic memory allocation path Unit testing GP Distance function Automatic bug fixing Structural Model Based GA ACO Test data generation for OO software Automatic test sequence generation data-flow (d-u) all-state feasibility Problem faced Source tended to avoid the branches that were hard to cover. Easy to detect faults may become harder to detect when they interact Manual target paths identification requires tester creativity, more time one parameter may not have effect on another function GP is computationall y expensive Girgis, Xiao [7] Lakhotia [8] Ahmed Hermadi[ 9] Alba Chicano[ 10] Arcuri[1 1] Liaskos[ 12] Li et al.[13]

5 Software through Evidence Gathering 629 Mutation Temporal Regression testing using slicing Integration testing software regression at the integration level. Unit testing Regression testing Genetic Programmi ng Evolutiona ry Algorithm Manual Generate evaluate test cases for the mutation testing. Verifying worst/best case execution time Coverage-focused, slicing Procedural-design firewall Data flow based Modificationfocused, minimization, branch bound algorithm Generatio n of test data to kill mutants Optimise worst/best case execution time GP is computationall y expensive EA alone is not sufficient for a thorough comprehensive test of real time systems. Emer Vergilio[ 14] Pohlheim Wegener [15] Gupta [16] Leung White [17] Harrold Soffa [18] Fischer, Hartman Robson [19] Thus a comparative analysis of the techniques used the testing strategies used reflect that if more efforts are done on path then White box testing can prove to be very effective. References [1] V. R. Basili R. W. Selby, "Comparing the effectiveness of software testing strategies," IEEE Trans. Software Eng., Vol. Se-13, No. 12, December [2] Bergstra, J.A.: About Instruction Sequence. arxiv: v1 [cs.se], 18 Jan [3] King, J.C.: Symbolic execution program testing. Communications of the ACM 19 (7), , 1976

6 630 Priyanka Mathur et al [4] Singh, Y.: Software. Cambridge University Press, Delhi, India, (2012), ISBN [5] Mark Harman, S. Afshin Mansouri Yuanyuan Zhang, Search Based Software Engineering: A Comprehensive Analysis Review of Trends Techniques Applications. Technical Report TR-09-03, April 9, 2009 [6] Mohd. Ehmer Khan, Different Forms of Software Techniques for Finding Errors. IJCSI International Journal of Computer Science Issues, Vol. 7, Issue 3, No 1, May 2010 [7] Girgis, M. R. (2005). Automatic Test Data Generation for Data Flow using a Genetic Algorithm. Journal of Universal Computer Science, 11(6), AND Xiao, M., El-Attar, M., Reformat, M., Miller, J. (2007). Empirical Evaluation of Optimization Algorithms when used in Goal-oriented Automated Test Data Generation Techniques. Empirical Software Engineering, 12(2), [8] Harman, M., Lakhotia, K., McMinn, P. (2007a). A Multi-Objective Approach to Search-based Test Data Generation. In Proceedings of the 9th annual Conference on Genetic Evolutionary Computation (GECCO 07), pages , London, Engl. ACM. [9] Ahmed, M. A. Hermadi, I. (2008). GA-based Multiple Paths Test Data Generator. Computers & Operations Research, 35(10), [10] Alba, E. Chicano, F. (2008). Observations in using Parallel Sequential Evolutionary Algorithms for Automatic Software. Computers & Operations Research, 35(10), [11] Arcuri, A. (2008). On the Automation of Fixing Software Bugs. In Proceedings of the Doctoral Symposium of the IEEE International Conference on Software Engineering (ICSE 08), pages , Leipzig, Germany. ACM. [12] Liaskos, K., Roper, M., Wood, M. (2007). Investigating Data-Flow Coverage of Classes Using Evolutionary Algorithms. In Proceedings of the 9th annual Conference on Genetic Evolutionary Computation (GECCO 07), 70 pages , London, Engl. ACM. [13] Li, H. Lam, C. P. (2005a). An Ant Colony Optimization Approach to Test Sequence Generation for Statebased Software. In Proceedings of the 5th International Conference on Quality Software (QSIC 05), pages , Melbourne, Australia. IEEE Computer Society. [14] Emer, M. C. F. P. Vergilio, S. R. (2002). GPTesT: A Tool Based On Genetic Programming. Proceedings of the 2002 Conference on Genetic Evolutionary Computation (GECCO 02), pages , New York, USA. Morgan Kaufmann Publishers.

7 Software through Evidence Gathering 631 [15] Pohlheim, H. Wegener, J. (1999). the Temporal Behavior of Real- Time Software Modules using Extended Evolutionary Algorithms. Proceedings of the Genetic Evolutionary Computation Conference (GECCO 99), volume 2,page 1795, Orlo, Florida, USA. Morgan Kaufmann. [16] Gupta, R., Harrold, M.J., Soffa, M.L An approach to regression testing using slicing. In Conference on Software Maintenance 1992 (Cat.No.92CH3206-0). IEEE Comput. Soc. Press, [17] Leung, H.K.N. White, L A study of integration testing software regression at the integration level. In Proceedings. Conference on Software Maintenance 1990 (Cat.No.90CH2921-5). IEEE Comput. Soc. Press, [18] Harrold, M.J. Souffa, M.L An incremental approach to unit testing during maintenance. In Proceedings of the Conference on Software Maintenance (IEEE Cat. No. 88CH2615-3). IEEE Comput. Soc. Press, [19] Hartmann, J. Robson, D.J Techniques for selective revalidation. IEEE Software.7(1),31-6.

8 632 Priyanka Mathur et al